Deep Neural Networks with Multi-Branch Architectures Are Intrinsically Less Non-Convex.
Hongyang ZhangJunru ShaoRuslan SalakhutdinovPublished in: AISTATS (2019)
Keyphrases
- neural network
- convex optimization
- pattern recognition
- artificial neural networks
- neural architectures
- back propagation
- fuzzy logic
- genetic algorithm
- convex hull
- neural network model
- multilayer perceptron
- piecewise linear
- data sets
- case study
- machine learning
- multi layer
- fuzzy systems
- network architecture
- globally optimal
- convex sets